A small population of neurons (~ 5000 cells) in the arcuate region of hypothalamus that make neuropeptide Y (NPY) and agouti-related protein (AgRP) become essential for survival in adult mice. We targeted the diphtheria toxin receptor (DTR) to these NPY/AgRP neurons to allow their ablation by administration of diphtheria toxin (DT). Ablation of these neurons in adult mice results in starvation. However, their ablation in neonatal mice results in circuit-based compensation such that mice grow almost normally. Although neonatally DT-lesioned mice grow to normal size, we experimentally explore whether they manifest deficits in physiological responses that normally depend on NPY/AgRP neurons. We propose to determine the critical signaling molecules made by NPY/AgRP neurons and whether they function primarily by regulating the melanocortin-signaling pathway or other circuits. A prime candidate for the critical signaling molecule is ?-aminobutyric acid (GABA);thus, we propose to make conditional alleles of the two genes, GAD1 and GAD2 encoding the biosynthetic enzymes such that they can be inactivated in NPY/AgRP neurons and temporally controlled manner by the action of Cre recombinase. If GABA is the critical signaling molecule, then we predict that inactivation of GABA biosynthesis in NPY/AgRP neurons of neonatal mice will lead to compensation such that the cells are no longer necessary for survival. In contrast, inactivation of GABA biosynthesis in NPY/AgRP neurons of adult mice may promote starvation. NPY/AgRP neurons are known to inhibit neighboring neurons in the arcuate nucleus that make proopiomelanocortin (POMC) and their target neurons in the paraventricular nucleus that express melanocortin-4 receptor. Activation of POMC neurons inhibits feeding. Thus, we include experiments to directly test whether up-regulation of this melanocortin-signaling pathway after sudden ablation of NPY/AgRP neurons is critically involved in the starvation phenotype. However, because NPY/AgRP neurons also project to several other brain regions, we include experiments that address which of these projection regions are affected the most and whether they contribute to feeding behavior. These experiments should provide insight into the neural circuits and signaling molecules that are critical for feeding.